Avalanche: adaptation of parallel and distributed computing for dynamic analysis to improve performance of defect detection

Abstract

This paper focuses on dynamic program analysis optimization through the use of distributed computing scheme and parallel computing for checking satisfiability of Boolean constraint sets. An overview of results obtained from applying the practical implementation of parallel and distributed schemes of dynamic analysis to a number of open-source applications is given. The paper presents a detailed evaluation of the increased efficiency of dynamic analysis achieved while applying developed techniques. Finally, the authors propose a number of possible directions for future work.

Keywords

• динамический анализ
• обнаружение дефектов
• параллельные вычисления
• распределенные вычисления